Non-woven fabric processing optimization method and system based on production data

Abstract

The invention relates to the technical field of production monitoring, in particular to a non-woven fabric processing optimization method and system.The method comprises the steps that firstly, a surface image of melt-blown non-woven fabric is obtained and processed, a non-woven fabric quality index is obtained, and if the surface image is larger than a preset non-woven fabric quality index threshold value, production is normal; if the non-woven fabric quality index is not larger than the non-woven fabric quality index threshold value, production data in the production process are obtained, a difference index is obtained by combining preset production parameters, and the difference index is divided into a first-level difference, a second-level difference and a third-level difference according to a difference index grading function; if the difference is the first-level difference, the production rate is controlled by improving a PID algorithm; if the speed adjustment fails, production safety early warning is carried out; if the difference is the second-level difference, performing staged rate control through an improved PID (Proportion Integration Differentiation) algorithm; and if the difference is the three-level difference, early warning that the production machine may have a fault. By monitoring and controlling the production process of the melt-blown non-woven fabric, the production efficiency is improved, and the stability of the production process is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of production monitoring, in particular to a non-woven fabric processing optimization method and system based on production data. BACKGROUND

[0002] Non-woven fabric is a kind of fabric made by combining fibers through physical or artificial methods. Unlike traditional woven fabric, it is usually produced by processes such as water jet, melt blowing and needle punching, which bond short fibers or filaments together without the need for weaving or knitting. Due to its lightness, breathability, durability, waterproofness, simple process, low cost and high production efficiency, it has wide application in medical and health care, home life and flame-retardant materials.

[0003] Melt-blown non-woven fabric is a kind of non-woven fabric material manufactured by melt-blown process, and its main raw material is thermoplastic polymer such as polypropylene. This material is widely used in filtration, medical protection and environmental protection due to its unique microfiber structure. The production of melt-blown non-woven fabric is a continuous process from polymer raw material to finished non-woven fabric, which needs to go through multiple process steps, including raw material melting, extrusion and spinning, fiber stretching and web reinforcement, etc.

[0004] The production process of melt-blown non-woven fabric requires precise process control to ensure efficient production and quality consistency. The production line speed control is crucial in the production process of melt-blown non-woven fabric, which ensures the stability and efficiency of the entire production process, and guarantees the stability of the connection between different steps in the production process. If the spinning and winding steps are not synchronized, it will lead to an increase in waste and a decrease in efficiency.

[0005] Therefore, a non-woven fabric processing optimization method and system based on production data are proposed. SUMMARY

[0006] The purpose of the present application is to provide a non-woven fabric processing optimization method and system based on production data. First, the surface image of melt-blown non-woven fabric is obtained and processed to obtain the non-woven fabric quality index, realizing rapid quality detection of melt-blown non-woven fabric on the production line. If the production quality is unqualified, the data in the production process are obtained and processed to obtain the difference index, which is used to check and evaluate the entire production process. Then, the production process is processed according to the difference index, and the improved PID algorithm is used to control the production parameters. If production problems are found, an early warning is sent in time.

[0007] To achieve the above purpose, the present application provides the following technical solutions:

[0008] A non-woven fabric production and processing optimization control method, comprising:

[0009] Step S1: obtaining an image of the melt-blown non-woven fabric and performing quality analysis to obtain a fiber thinning index, a uniformity index, and a non-woven fabric quality index; determining whether to perform step S2 according to the fiber thinning index, the uniformity index, and the non-woven fabric quality index;

[0010] Further, the non-woven fabric quality index comprises:

[0011] Obtaining a surface image of the melt-blown non-woven fabric at a fixed interval and processing the image to obtain image data; obtaining the image data and processing the image data to obtain a fiber thinning index; if the fiber thinning index is not greater than a preset fiber thinning index threshold value, it indicates that there is a problem in production, and step S2 is performed.

[0012] Obtaining a light transmission image of the melt-blown non-woven fabric and processing the image to obtain light transmission data; obtaining the light transmission data and processing the light transmission data to obtain a uniformity index; if the uniformity index is not greater than a preset uniformity index threshold value, it indicates that there is a problem in production, and step S2 is performed.

[0013] If the fiber thinning index is greater than the preset fiber thinning index threshold value and the uniformity index is greater than the preset uniformity index threshold value, the non-woven fabric quality index is calculated; if the non-woven fabric quality index is greater than a preset non-woven fabric quality index threshold value, it indicates that the production is normal; if the non-woven fabric quality index is not greater than the preset non-woven fabric quality index threshold value, it indicates that there is a problem in production, and step S2 is performed.

[0014] Step S2: obtaining production data in the production process of the melt-blown non-woven fabric, the production data comprising: melt extrusion rate, jetting rate, air flow stretching rate, and collection web rate.

[0015] Further, the production data further comprises:

[0016] Calculating the output per unit time of each step in the production process of the melt-blown non-woven fabric, and calculating the error value between the output per unit time of each step to obtain output error data.

[0017] Step S3: obtaining the production data and preset production parameters and processing to obtain a difference index; according to a difference index classification function, the difference index is classified into a first-level difference, a second-level difference, and a third-level difference.

[0018] Further, the difference index comprises:

[0019] acquiring the production data and processing to obtain the difference index; if the difference index is not greater than a first difference index threshold, it is a first-level difference, and step S4 is executed; if the difference index is greater than the first difference index threshold and not greater than a second difference index threshold, it is a second-level difference, and step S5 is executed; if the difference index is greater than the second difference index threshold, it is a third-level difference, and step S6 is executed; the first difference index threshold is less than the second difference index threshold.

[0020] Step S4: if it is the first-level difference, the production data and the preset production parameter are acquired and production rate control is performed through the improved PID algorithm; if rate adjustment fails, production safety warning is performed.

[0021] Further, the improved PID algorithm comprises:

[0022] The coefficients of the improved PID algorithm are optimized by using a genetic algorithm, and the coefficients include a proportional coefficient, an integral coefficient and a differential coefficient.

[0023] The improved PID algorithm obtains a result by adding a proportional term, an integral term and a differential term, wherein the proportional term is a value obtained by multiplying a proportional coefficient by a difference between an error at a current time and an error at a previous time; the integral term is a value obtained by multiplying an integral coefficient by the error at the current time; and the differential term is a value obtained by multiplying a differential coefficient by the error at the current time, subtracting twice the error at the previous time, and adding the error at the previous time.

[0024] Further, the production rate control comprises:

[0025] A difference value set is obtained by acquiring the to-be-controlled parameters and the preset production parameter and processing; the to-be-controlled parameters are prioritized according to the difference values in the difference value set to obtain sorted to-be-controlled parameters; and the sorted to-be-controlled parameters are acquired according to the priorities and controlled according to the improved PID algorithm.

[0026] Further, the rate adjustment failure comprises:

[0027] A first parameter control duration is acquired, the first parameter control duration representing a complete time for which a to-be-controlled parameter is controlled according to the improved PID algorithm; if the first parameter control duration is greater than a preset first parameter control duration, it indicates that rate adjustment fails, and production safety warning is performed.

[0028] Step S5: if it is the second-level difference, the production data are acquired and processed to obtain a to-be-controlled rate increment, and staged rate control is performed through the improved PID algorithm; if rate adjustment fails, production safety warning is performed.

[0029] Further, the staged rate control comprises:

[0030] The to-be-controlled parameter and the preset production parameter are acquired and processed to obtain a first to-be-controlled rate increment; the first to-be-controlled rate increment is segmented to obtain N second to-be-controlled rate increments; the second to-be-controlled rate increment is acquired and rate control is performed according to the improved PID algorithm;

[0031] Further, the time of the N staged rate controls is acquired and processed to obtain a second parameter control time length; if the second parameter control time length is greater than a preset second parameter control time length, it indicates that rate adjustment fails, and production safety warning is performed.

[0032] Step S6: If it is the third-level difference, it indicates that the production machine has a fault, and production safety warning is performed.

[0033] Further, the production safety warning comprises: if it is the third-level difference, it indicates that there is a production problem in the entire non-woven fabric production and processing process, and a warning is issued;

[0034] Further, the to-be-controlled parameter and data of the adjustment failure are acquired, and the processing step and position of the production problem are warned.

[0035] The application also provides a non-woven fabric production and processing optimization system, comprising:

[0036] A non-woven fabric quality analysis module is used to acquire melt-blown non-woven fabric and perform quality analysis to obtain a non-woven fabric quality index; if the non-woven fabric quality index is greater than a preset non-woven fabric quality index threshold, it indicates that the production is normal; if it is not greater than the non-woven fabric quality index threshold, it indicates that there is a production problem, and the non-woven fabric production parameter acquisition module is entered;

[0037] A non-woven fabric production parameter acquisition module is used to acquire production data in the melt-blown non-woven fabric production process; the production data comprises: melt extrusion rate, jet rate, air flow stretching rate and collection web rate;

[0038] A difference index calculation module is used to acquire the production data and preset production parameter and process to obtain a difference index, and the difference index is classified into a first-level difference, a second-level difference and a third-level difference according to a difference index classification function;

[0039] A first-level difference processing module is used to acquire the production data and the preset production parameter and perform production rate control through an improved PID algorithm; if rate adjustment fails, production safety warning is performed;

[0040] A secondary difference processing module is configured to acquire the production data and process the data to obtain a to-be-controlled rate increment, and control the rate in stages through the improved PID algorithm; if the rate adjustment fails, a production safety warning is given.

[0041] A tertiary difference processing module is configured to give a warning that production needs to be stopped and the machine needs to be checked.

[0042] Compared with the prior art, the present application has the following advantages:

[0043] 1. The surface image and the light transmission image of the melt-blown non-woven fabric on the production line are collected at fixed intervals and processed to obtain the fiber thinning index and the uniformity index, and further obtain the non-woven fabric quality index, so that real-time quality detection of the melt-blown non-woven fabric on the production line can be realized, high-quality production can be ensured, and the quality detection efficiency of the melt-blown non-woven fabric can be improved; in the case that the non-woven fabric quality index is not greater than a preset non-woven fabric quality index threshold, the parameters in the production process are collected and processed, and data analysis is performed to provide data support for subsequent parameter adjustment and control.

[0044] 2. By analyzing the melt extrusion rate, the spinning rate, the air flow stretching rate, the collection web rate and other key production parameters, the state of each link in the production process can be accurately controlled; by calculating the predicted output of each production link and analyzing the error between the predicted outputs, abnormalities in the production process can be timely reflected, which helps to accurately adjust the process parameters, so as to ensure the matching and cooperation of each link; through the hierarchical response of the difference index, the system can take corresponding control measures according to the severity of the problem.

[0045] 3. If it is a first-level difference, the improved PID algorithm is used to adjust and control the production parameters, the production rate of each process can be adjusted in real time and accurately, and the downtime and maintenance cost can be reduced; if it is a second-level difference, the to-be-controlled increment is segmented first and then controlled in stages; if it is a third-level difference, it indicates that there is a problem in the production process, and a safety warning is given; through the rate fine-tuning of the first-level difference, the stage-by-stage control of the second-level difference and the direct warning of the third-level difference, production problems can be timely fed back and handled, and the efficiency and safety of the overall production are improved. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 A flowchart of a non-woven fabric processing optimization method based on production data provided for the first embodiment of the present application;

[0047] Figure 2 A flowchart of a non-woven fabric quality index calculation method in the first embodiment of the present application;

[0048] Figure 3A structure schematic diagram of a non-woven fabric processing optimization system based on production data provided for Embodiment 2 of the present application. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0050] Embodiment 1

[0051] A textile company introduced a non-woven fabric processing optimization method based on production data provided by the present application in the production process of melt-blown non-woven fabric to improve the production efficiency of melt-blown non-woven fabric and ensure the smoothness and efficiency of the entire melt-blown non-woven fabric production process. The specific process of the method is as shown in Figure 1 The specific implementation is as follows:

[0052] Step S1: obtaining an image of melt-blown non-woven fabric and performing quality analysis to obtain a fiber thinning index, a uniformity index and a non-woven fabric quality index; determining whether to execute step S2 according to the fiber thinning index, the uniformity index and the non-woven fabric quality index;

[0053] Further, the calculation process of the non-woven fabric quality index is as shown in Figure 2 , which includes:

[0054] Collecting surface images of melt-blown non-woven fabric at fixed intervals and processing to obtain image data; obtaining the image data and processing to obtain a fiber thinning index; if the fiber thinning index is not greater than a preset fiber thinning index threshold, it indicates that there is a problem in production, and step S2 is executed;

[0055] Further, the surface images of the melt-blown non-woven fabric are processed for denoising and enhancement to obtain surface images with more obvious contrast; and then the surface images are uniformly segmented to obtain a segmented image set;

[0056] Further, a number of segmented images are randomly obtained from the segmented image set, and the approximate diameters of fibers are extracted from each image by setting a preset pixel value range. The fiber thinning index is calculated as follows: first, the average value of all fiber approximate diameters extracted from the segmented images is calculated; second, the absolute difference between the average value and a preset fiber diameter is obtained; then, the difference is multiplied by a preset magnification coefficient (10 3 in this embodiment); finally, the calculation result of the previous step is taken as a negative value, and the exponential function value is obtained to obtain the fiber thinning index. 3 in this embodiment); finally, the calculation result of the previous step is taken as a negative value, and the exponential function value is obtained to obtain the fiber thinning index.

[0057] acquire a light transmission image of the melt-blown nonwoven fabric under illumination and process it to obtain light transmission data, and calculate a uniformity index; if the uniformity index is not greater than a preset uniformity index threshold, it indicates that there is a problem in production, and step S2 is executed;

[0058] Further, a uniform light source is arranged above the melt-blown nonwoven fabric, a light transmission image thereof is acquired, and after processing such as denoising and enhancement, a clear gray-scale light transmission image is obtained, and the image is uniformly divided into a plurality of segmented images. The calculation of the uniformity index comprehensively considers both macro-uniformity and micro-defects. To obtain a value representing macro-uniformity, the absolute difference between the average gray-scale value of each segmented image and the average gray-scale value of the whole image is calculated and summed up, then 1 is subtracted from the quotient of the sum divided by the product of the number of segmented images and the average gray-scale value of the whole image, and the result is multiplied by a weight coefficient of the average gray-scale value. At the same time, to obtain a value representing micro-defects, the total number of pixel points with a gray-scale value not in a preset range in all segmented images is counted, and multiplied by a weight coefficient of the number of pixel points. Finally, the value representing macro-uniformity is subtracted from the value representing micro-defects, and the final uniformity index is obtained.

[0059] When both the fiber thinning index and the uniformity index are greater than the respective preset thresholds, the system will further calculate the nonwoven fabric quality index. The calculation method of the quality index is as follows: the fiber thinning index is multiplied by its corresponding weight coefficient, the uniformity index is multiplied by its corresponding weight coefficient, and finally the two products are added to obtain the final result. The calculated quality index is compared with a preset quality index threshold, if it is greater than the threshold, it indicates that the production process is normal; if it is not greater than the threshold, it indicates that there may be a problem in production, and the subsequent inspection step S2 needs to be performed. The weight coefficient is obtained by collecting a plurality of batches of nonwoven fabric samples, and measuring the fiber thinning index, the uniformity index and the preset final quality performance index (such as filtration efficiency or breaking strength) of each sample; then, the final quality performance index is used as the dependent variable, and the fiber thinning index and the uniformity index are used as the independent variables, and multiple linear regression analysis is performed; finally, the regression coefficients of the respective independent variables obtained by regression analysis are normalized to obtain the weight coefficients corresponding to the fiber thinning index and the uniformity index, respectively.

[0060] Table 1, nonwoven fabric quality index at some time nodes

[0061] Time node Fiber refinement index Uniformity index Nonwoven mass index Is greater than threshold Node 1 0.85 0.91 0.88 Yes Node 2 0.92 0.91 0.92 Yes Node 3 0.92 0.88 0.90 Yes Node 4 0.74 0.81 0.78 No

[0062] Table 1 shows the results of quality detection of the melt-blown non-woven fabric at some time points. It can be found that the melt-blown non-woven fabric on the production line is unqualified when the fourth quality detection is performed, indicating that there may be a production problem. By collecting the surface image and light transmission image of the melt-blown non-woven fabric on the production line at fixed intervals and processing, the fiber refinement index and uniformity index are obtained, and the non-woven fabric quality index is further obtained, realizing the rapid evaluation and detection of the quality of the melt-blown non-woven fabric on the production line, ensuring high-quality production, and improving the quality detection efficiency of the melt-blown non-woven fabric; in the case that the non-woven fabric quality index is not greater than the preset non-woven fabric quality index threshold, it indicates that there may be a problem in the production process, and the detection efficiency of the production problem is improved.

[0063] Step S2: acquiring production data in the melt-blown non-woven fabric production process, the production data including: melt extrusion rate, jetting rate, air flow stretching rate and collection net rate;

[0064] Further, the production data further includes a series of parameters such as the rotation speed of the extruder screw, the melt temperature, the jetting plate temperature, the air flow temperature, the air flow pressure, the temperature and flow rate of the cooling air, the fiber cooling distance and the winding tension.

[0065] Further, all production parameters and preset production parameters are acquired and normalized and standardized to obtain production parameter vectors and preset production parameter vectors, and the production parameter error between the production parameters and the preset production parameters is calculated by using cosine similarity.

[0066] Further, the production data further includes:

[0067] The above production data within a unit time is acquired and the actual output within a unit time of each step is calculated, such as the actual output of the melt extrusion process, the jetting process and the collection net collection process, and then the error value between the actual outputs of each step is calculated to obtain output error data, the output error data including the output error Error1 between the melt extrusion process and the jetting process and the output error Error2 between the jetting process and the collection net collection process.

[0068] Through comprehensive data collection, a complete production data vector is formed, and compared with the preset production parameters, each production link can be accurately monitored in real time, which is helpful to realize high-quality production control; by collecting data and calculating output error data, whether the entire melt-blown non-woven fabric production process is stable and consistent can be measured, providing data support for subsequent parameter control.

[0069] Step S3: acquiring the production data and the preset production parameters and processing to obtain a difference index; according to a difference index grading function, the difference index is divided into a first-level difference, a second-level difference and a third-level difference;

[0070] Specifically, after obtaining the production parameter error and output error data, the difference index is calculated by adding three independent weighted error terms. The first error term represents the deviation of the overall production parameters, which is calculated by subtracting the cosine similarity between the real-time production parameter vector and the preset production parameter vector from 1, and then multiplying the result by its corresponding weight coefficient; the second error term is the output error between the melt extrusion process and the spinning process, multiplied by its own weight coefficient; the third error term is the output error between the spinning process and the collection process of the collection net, also multiplied by its own weight coefficient. Adding the three calculated weighted error terms together gives the final difference index. Then, according to the value of the difference index, the classification process is carried out: if the index is not greater than the first difference index threshold, it is a first-level difference, step S4 is executed; if it is greater than the first threshold but not greater than the second threshold, it is a second-level difference, step S5 is executed; if it is greater than the second threshold, it is a third-level difference, step S6 is executed, and the first difference index threshold is less than the second difference index threshold.

[0071] Step S4: If it is the first-level difference, obtain the production data and the preset production parameters and control the production rate by improving the PID algorithm; if the rate adjustment fails, a production safety warning is given;

[0072] Further, the improved PID algorithm includes:

[0073] Further, the coefficients of the improved PID algorithm, including the proportional coefficient, the integral coefficient and the differential coefficient, are optimized by using a genetic algorithm.

[0074] The improved PID algorithm adds the proportional term, the integral term and the differential term to obtain the result, wherein the proportional term is the value obtained by multiplying the proportional coefficient by the difference between the error at the current time and the error at the last time; the integral term is the value obtained by multiplying the integral coefficient by the error at the current time; the differential term is the value obtained by multiplying the differential coefficient by the error at the current time minus twice the error at the last time, plus the error at the two times.

[0075] Further, during the simulation training, the proportional coefficient, the integral coefficient and the differential coefficient of the PID are optimized by using a genetic algorithm. The genetic algorithm generates a set of parameter combinations, then calculates the performance indicators of each group of parameters, selects the better combinations for crossover and mutation, and finally finds the optimal solution.

[0076] Further, a set of random K p , K i and K d combinations are first generated, and then the PID controller is run to calculate the performance indicators of each group of parameters, which can use response time or steady-state error as performance indicators.

[0077] Further, the excellent combinations are then selected according to the fitness for cross and mutation to generate the next generation population; the iteration is repeated until the optimal parameter combination is found.

[0078] Further, the fitness function is designed by calculating the weighted combination of multiple performance indicators. The specific calculation method is as follows: the response time is multiplied by its corresponding weight coefficient, the overshoot is multiplied by its corresponding weight coefficient, and the steady-state error is multiplied by its corresponding weight coefficient. Finally, the three calculated products are added together, and the resulting sum is the final fitness function value.

[0079] When determining the weight coefficients of each performance indicator in the fitness function, the setting depends on the performance focus preset by the controlled system. These weight coefficients, i.e., the weights of response time, overshoot, and steady-state error, reflect the importance trade-off between different performance indicators, guiding the optimization direction of the optimization algorithm.

[0080] Specifically, the following principles can be used for setting:

[0081] If the system prioritizes fast response, priority should be given to reducing response time. At this time, a relatively high value will be assigned to the weight coefficient of response time, while the weight coefficients of overshoot and steady-state error will be adjusted lower.

[0082] If the system prioritizes stability to ensure process smoothness and safety, priority should be given to reducing overshoot. At this time, the highest value will be assigned to the weight coefficient of overshoot, while the weight coefficients of response time and steady-state error will be set lower.

[0083] If the system prioritizes final accuracy, pursuing the minimum error after long-term operation, priority should be given to reducing steady-state error. At this time, the highest value will be assigned to the weight coefficient of steady-state error, and the weights of response time and overshoot will be reduced accordingly.

[0084] In practical applications, the final determination of these weight coefficients is usually combined with the above engineering requirement analysis and simulation testing, by adjusting the size of each weight to balance the speed, stability, and accuracy of the system, so that the performance of the optimized PID controller best meets the specific application scenario.

[0085] Further, the response time represents the time required for the system to rise from a certain initial value to its steady-state value, and a short rise time means that the system can respond quickly; the overshoot represents the maximum difference between the system control process and the steady-state value, usually expressed as a percentage, that is, the degree to which the response signal exceeds the target value before reaching the final target value; the steady-state error represents the difference between the actual output value and the expected output value after the system reaches steady state, and is used to measure the accuracy that the system can ultimately achieve.

[0086] Further, after selecting the optimal proportional coefficient, integral coefficient and differential coefficient through the genetic algorithm, they are set as initial coefficients, and a set of dynamic coefficient rules is designed to adjust these three parameters in real time based on the error size. The specific adjustment method is as follows:

[0087] At any time, the new proportional coefficient is obtained by adding the product of the absolute value of the current error and a gain adjustment coefficient to the initial proportional coefficient.

[0088] Similarly, the new integral coefficient is obtained by adding the product of the absolute value of the current error and another gain adjustment coefficient to the initial integral coefficient.

[0089] Similarly, the new differential coefficient is obtained by adding the product of the absolute value of the current error and a third gain adjustment coefficient to the initial differential coefficient.

[0090] By introducing the improved PID algorithm control and adjusting the production parameters, more efficient and accurate control effect can be achieved; first, the incremental PID control algorithm is more stable than the traditional PID algorithm, and has stronger anti-interference ability for the control system, second, the genetic algorithm can find the optimal parameter combination through the evaluation and selection of the fitness function, avoiding falling into a local optimal solution, and finally, the coefficients of the algorithm are adjusted according to the real-time error size, making the control system more flexible to adapt to different production conditions and environmental changes, and enhancing the adaptive ability of the system.

[0091] Further, the production rate control includes:

[0092] The parameters to be controlled and the corresponding preset production parameters are obtained and processed to obtain a difference set; the data in the difference set are used to prioritize all the parameters to be controlled to obtain sorted parameters to be controlled; the parameters to be controlled are obtained according to the priority, and the production parameter control is performed according to the improved PID algorithm.

[0093] Further, the to-be-controlled parameters include melt extrusion rate, jetting rate, air flow stretching rate and collection web rate, the preset production parameters corresponding thereto are acquired and absolute errors are calculated to obtain a difference set {R}, and the absolute errors are sorted to obtain sorted to-be-controlled parameters. Finally, the to-be-controlled parameters are acquired according to priorities and rate control is performed through the improved PID algorithm. As shown in Table 2, specific data of some to-be-controlled parameters are shown.

[0094] Table 2, some parameters

[0095] Parameter to be controlled Preset production parameter Melt extrusion rate (g / hole / min) 0.07 0.06 Roller rotation speed (m / min) 92.41 94.00

[0096] By acquiring the differences between the to-be-controlled parameters and the preset production parameters and performing priority sorting, rate control is performed in combination with the improved PID algorithm, the production parameter control process is optimized, and efficient, accurate and flexible control of the production process is realized.

[0097] Further, the rate adjustment failure in step S4 includes:

[0098] A first parameter control duration is acquired, which represents a complete duration of parameter adjustment of the to-be-controlled parameter according to the improved PID algorithm. If the first parameter control duration is greater than a preset first parameter control duration, it indicates that the rate adjustment fails, and a production safety warning is performed.

[0099] By monitoring the adjustment time of the to-be-controlled parameter and comparing it with the preset first parameter control duration, the to-be-controlled parameter with rate adjustment failure can be identified in time, and a warning can be issued in time, reducing potential risks and further ensuring the stability, safety and efficiency of the production process.

[0100] Step S5: If the second difference is obtained, the production data are acquired and processed to obtain a to-be-controlled rate increment, and the improved PID algorithm is used for stage-by-stage rate control. If the rate adjustment fails, a production safety warning is performed.

[0101] Further, the stage-by-stage rate control includes:

[0102] The to-be-controlled parameters and corresponding preset parameters are acquired and processed to obtain a first to-be-controlled rate increment. The to-be-controlled parameters include melt extrusion rate, jetting rate, air flow stretching rate and collection web rate. The to-be-controlled rate increment is uniformly divided N-1 times to obtain N second to-be-controlled rate increments. The second to-be-controlled rate increments are acquired and N times of rate control is performed according to the improved PID algorithm.

[0103] Further, a second parameter control duration of the N stage rate control is obtained, and if the second parameter control duration is greater than a preset second parameter control duration, it indicates that the rate adjustment fails, and a production safety warning is given.

[0104] The secondary difference is controlled in stages, the rate increment is divided into multiple stages, and the improved PID algorithm is used for control, so that the production rate can be flexibly adjusted, the stability of the control system is improved, and the risk of system instability or production accidents caused by excessive adjustment is reduced; in each control stage, the second parameter control duration is monitored in real time, and if the rate adjustment fails, a production safety warning can be triggered in time to improve the production safety.

[0105] Step S6: If the third difference is detected, it indicates that the production machine may have a fault, and a production warning is given.

[0106] Further, if the difference index indicates the third difference, it indicates that there is a real-time production problem in the entire non-woven fabric production process, and a warning is given to stop production and check the machine.

[0107] Further, the production safety warning in steps S4 and S5 includes:

[0108] The failed control parameter and its related data are obtained, and the processing step with a production problem and the machine position corresponding to the control parameter are warned.

[0109] In step S6, if the third difference is detected and a production warning is given, it means that the production machine may have a fault and needs to be checked; through the difference index and its related error data, the serious problem existing in the production process can be quickly identified, the production machine position where the fault may occur can be located, the stability of the melt-blown non-woven fabric production is enhanced, the production process is improved, and the production efficiency is improved.

[0110] The non-woven fabric processing optimization method based on production data provided by the application first judges whether there is a problem in the production process by detecting the quality of the produced melt-blown non-woven fabric, realizes the rapid detection of the production quality, if the quality of the melt-blown non-woven fabric is unqualified, it is processed in stages, if it is a first and second difference, the parameters in the production process are obtained, and the improved PID algorithm is used for parameter control, if it is a third difference, it indicates that there is a big problem in the production process of the melt-blown non-woven fabric, and the production needs to be stopped and the machine needs to be checked. Through the method provided by the application, the production efficiency and stability of the melt-blown non-woven fabric are improved.

[0111] Example 2

[0112] Flocking is a process that makes the flocking hair with negative charge and then vertically fixes it on the substrate coated with adhesive by the physical property of same charge repulsion and different charge attraction; non-woven fabric is a sheet or web formed by mutual adhesion of fibers through physical or chemical methods. A cloth manufacturer combines non-woven fabric and flocking to use non-woven fabric as the bottom fabric of flocking, and tries an innovative technology of applying flocking process to the surface of non-woven fabric. In order to produce non-woven fabric meeting the flatness and cleanliness requirements of flocking bottom fabric, the cloth manufacturer introduces a non-woven fabric processing optimization method system based on production data provided by the present application, the system structure is shown in Figure 3 as follows, comprising:

[0113] The non-woven fabric quality analysis module is used to obtain melt-blown non-woven fabric and perform quality analysis to obtain a non-woven fabric quality index. If the non-woven fabric quality index is greater than a preset non-woven fabric quality index threshold, it indicates that the production is normal, and if the non-woven fabric quality index is not greater than the preset non-woven fabric quality index threshold, it indicates that there is a problem in the production, and the non-woven fabric production parameter acquisition module is entered.

[0114] Further, the surface image of the melt-blown non-woven fabric is collected at intervals and processed to obtain image data. The image data is obtained and processed to obtain a fiber refinement index. If the fiber refinement index is not greater than a preset refinement index threshold, it indicates that there is a problem in the production, and the fiber diameter of the produced melt-blown non-woven fabric does not meet the requirements, and the non-woven fabric production parameter acquisition module is entered.

[0115] Further, the present embodiment provides another calculation method in the process of calculating the uniformity index. First, the surface image of the melt-blown non-woven fabric is obtained and processed to obtain a gray-scale image. The positions of the shadow pixel points in the gray-scale image are extracted. If the pixel value of the pixel point in the gray-scale image is less than a preset pixel value threshold, it is marked as a shadow pixel point. All pixel points are judged to obtain C shadow pixel point regions.

[0116] Further, the cth shadow pixel point region is obtained, and its geometric center is calculated to obtain the position of the shadow pixel point region center. The nearest other shadow pixel point region center in the shadow pixel point region center field is obtained and processed to obtain the nearest neighbor distance d c .

[0117] Further, after obtaining the nearest neighbor distances of all the shadow pixel point region centers, the distance data is processed by the following process to obtain the uniformity index: first, the average value and the standard deviation of all the nearest neighbor distances are calculated; then, the standard deviation is divided by the average value to obtain a ratio reflecting the uniformity of the distance distribution; then, the ratio is multiplied by a preset weight coefficient; finally, 1 is subtracted from the product obtained in the previous step, and the result is the uniformity index calculated according to the nearest neighbor distance. If the uniformity index is not greater than the preset uniformity index threshold, it indicates that there is a problem in production, and the non-woven fabric production parameter acquisition module is entered.

[0118] If the fiber thinning index is greater than the preset fiber thinning index threshold and the uniformity index is greater than the preset uniformity index threshold, the non-woven fabric quality index is calculated according to the fiber thinning index and the uniformity index; if the non-woven fabric quality index is greater than the preset non-woven fabric quality index threshold, it indicates that the quality is not a problem; if the non-woven fabric quality index is not greater than the preset non-woven fabric quality index threshold, it indicates that there is a problem in production, and the non-woven fabric production parameter acquisition module is entered. As shown in Table 3, the results of the melt-blown non-woven fabric quality detection at some time points can be seen that there is no quality problem in the production process of the melt-blown non-woven fabric.

[0119] Table 3, non-woven fabric quality index at some time points

[0120] Time node Fiber refinement index Uniformity index Nonwoven mass index Is greater than threshold Node A 0.95 0.91 0.93 Yes Node B 0.97 0.92 0.95 Yes Node C 0.94 0.91 0.93 Yes Node D 0.94 0.92 0.93 Yes

[0121] The non-woven fabric production parameter acquisition module is used to acquire production data in the production process of the melt-blown non-woven fabric, and the production data includes: melt extrusion rate, jetting rate, air flow stretching rate and collection net rate;

[0122] Further, the production data further includes output error data, and the output error data is acquired in the following manner:

[0123] The production data in a unit of time is acquired and the actual output in a unit of time of each step is calculated, such as the actual output in a unit of time of the melt extrusion process, the jetting process and the collection net collection process, and then the error value between the actual outputs of each step is calculated to obtain the output error data, which includes the output error between the melt extrusion process and the jetting process and the output error between the jetting process and the collection net collection process.

[0124] The difference index calculation module is used to acquire the production data and the preset production parameters and process them to obtain the difference index, and the difference index is classified into a first-level difference, a second-level difference and a third-level difference according to the difference index classification function.

[0125] Further, the production parameter vector and the preset production parameter vector are obtained, and a production parameter error between the production parameter and the preset production parameter is calculated by using cosine similarity; and then a difference index is obtained in combination with the output error data.

[0126] Further, if the difference index is not greater than a first difference index threshold, it is a first-level difference, and a first-level difference processing module is entered for production parameter control; if the difference index is greater than the first difference index threshold and not greater than a second difference index threshold, it is a second-level difference, and a second-level difference processing module is entered and production parameter control is performed; and if the difference index is greater than the second difference index threshold, it is a third-level difference, and a third-level difference processing module is entered. The first difference index threshold is less than the second difference index threshold, and the second difference index threshold is less than a third difference index threshold.

[0127] The first-level difference processing module is used for obtaining the production data and the preset production parameter and performing production rate control by using the improved PID algorithm; and if rate adjustment fails, production safety warning is performed.

[0128] Further, the to-be-controlled parameters include melt extrusion rate, spinning rate, air flow stretching rate and collection net rate, the preset production parameters corresponding thereto are obtained, absolute errors are calculated, a difference set is obtained, the to-be-controlled parameters are sorted according to the sizes of the differences to obtain sorted to-be-controlled parameters, and finally the to-be-controlled parameters are obtained according to their priorities and rate control is performed by using the improved PID algorithm mentioned above.

[0129] Further, a first parameter control time length is obtained, which represents a complete time for parameter adjustment of the to-be-controlled parameters according to the improved PID algorithm; if the first parameter control time length is greater than a preset first parameter control time length, it indicates that rate adjustment fails, and production safety warning is performed.

[0130] The second-level difference processing module is used for obtaining the production data and processing to obtain a to-be-controlled rate increment, performing stage-by-stage rate control by using the improved PID algorithm; and if rate adjustment fails, production safety warning is performed.

[0131] Further, the second-level difference processing module obtains the to-be-controlled parameters and corresponding preset parameters and processes to obtain a first to-be-controlled rate increment; then the to-be-controlled rate increment is segmented to obtain a second to-be-controlled rate increment; and the second to-be-controlled rate increment is obtained and stage-by-stage rate control is performed according to the improved PID algorithm.

[0132] Further, a complete control time length of stage-by-stage control is obtained and processed to obtain a second parameter control time length; and if the second parameter control time length is greater than a preset second parameter control time length, it indicates that rate adjustment fails, and production safety warning is performed.

[0133] Further, if the difference index represents a third-level difference, it indicates that the difference between the production parameters in the production process and the preset production parameters is too large, and then enters the third-level difference processing module, which is used to issue a warning that the production needs to be stopped and machine inspection is needed.

[0134] By applying the non-woven fabric production and processing optimization system provided by the application, real-time monitoring and adjustment of the melt-blown non-woven fabric production and processing process can be realized, the production line can be ensured to be in an optimal state at all times, production efficiency and stability can be improved, and high-quality non-woven fabric meeting the requirements of the flocking base cloth can be produced.

[0135] After the high-quality non-woven fabric is produced, in order to enable the flock to be firmly adhered to the non-woven fabric, an adhesive is coated on the surface of the non-woven fabric, and the type and coating amount of the adhesive need to be determined according to the material of the non-woven fabric and the requirements of the flocking; then, the flock is given a negative charge through high-voltage electrostatic generated by the flocking machine, and the flocking is performed on the surface of the non-woven fabric coated with the adhesive, under the action of the high-voltage electric field, the flock will be accelerated vertically to the surface of the non-woven fabric and be adhered to the non-woven fabric vertically.

[0136] Further, after the flocking is completed, the non-woven fabric needs to be dried and solidified to ensure that the adhesive is completely solidified and the flock is firmly adhered to the non-woven fabric. Through this combination, the non-woven fabric can obtain the soft and comfortable texture, three-dimensional effect and visual effect brought by the flocking, and has wide application in the fields of home textiles, toys and automotive interiors.

[0137] Although the embodiments of the application have been shown and described, it is to be understood that, for those skilled in the art, various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirit of the application, and the scope of the application is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing nonwoven fabric processing based on production data, characterized in that, include: Step S1: Acquire images of meltblown nonwoven fabric and perform quality analysis to obtain fiber refinement index, uniformity index and nonwoven fabric quality index; Determine whether to execute step S2 based on the fiber refinement index, the uniformity index, and the nonwoven fabric quality index; Step S2: Obtain production data during the production process of the meltblown nonwoven fabric, including: melt extrusion rate, spinning rate, airflow stretching rate, and collection net rate; Step S3: Obtain and process the production data and preset production parameters to obtain the difference index; divide the difference index into first-level difference, second-level difference and third-level difference according to the difference index classification function; Step S4: If it is the first-level difference, then acquire the production data and the preset production parameters and control the production rate through an improved PID algorithm; if the rate adjustment fails, then issue a production safety warning. Step S5: If it is the second-level difference, then acquire and process the production data to obtain the second rate increment to be controlled, and perform phased rate control through the improved PID algorithm; if the rate adjustment fails, then issue a production safety warning. Step S6: If it is the third level difference, it indicates that there is a fault in the production machine, and a production safety warning is issued.

2. The method for optimizing nonwoven fabric processing based on production data according to claim 1, characterized in that, Determining whether to execute step S2 includes: The surface images of the meltblown nonwoven fabric are acquired at fixed intervals and processed to obtain image data; the image data is acquired and processed to obtain the fiber refinement index; if the fiber refinement index is not greater than the preset fiber refinement index threshold, it indicates that there is a problem in production, and step S2 is executed; Obtain and process the light-transmitting image of the meltblown nonwoven fabric to obtain light-transmitting data. Obtain and process the light-transmitting data to obtain a uniformity index. If the uniformity index is not greater than a preset uniformity index threshold, it indicates that there is a problem in production, and step S2 is executed. If the fiber refinement index is greater than the preset fiber refinement index threshold and the uniformity index is greater than the preset uniformity index threshold, then the nonwoven fabric quality index is calculated; if the nonwoven fabric quality index is greater than the preset nonwoven fabric quality index threshold, then production is normal; if the nonwoven fabric quality index is not greater than the preset nonwoven fabric quality index threshold, then there is a problem with production, and step S2 is executed.

3. The method for optimizing nonwoven fabric processing based on production data according to claim 1, characterized in that, After acquiring the production data, output error data is also acquired, including: Calculate the output per unit time for each step in the production process of the meltblown nonwoven fabric, obtain and process the output per unit time, and obtain output error data.

4. The method for optimizing nonwoven fabric processing based on production data according to claim 1, characterized in that, The difference index includes: The production data is acquired and processed to obtain the difference index; if the difference index is not greater than the first difference index threshold, it is a level 1 difference, and step S4 is executed; if the difference index is greater than the first difference index threshold but not greater than the second difference index threshold, it is a level 2 difference, and step S5 is executed; if the difference index is greater than the second difference index threshold, it is a level 3 difference, and step S6 is executed; the first difference index threshold is less than the second difference index threshold.

5. The method for optimizing nonwoven fabric processing based on production data according to claim 1, characterized in that, The improved PID algorithm includes: The coefficients of the improved PID algorithm are optimized using a genetic algorithm, and the coefficients include proportional coefficient, integral coefficient, and derivative coefficient. The improved PID algorithm obtains the result by adding the proportional term, integral term, and derivative term. The proportional term is the value obtained by multiplying the proportional coefficient by the difference between the error at the current time and the error at the previous time. The integral term is the value obtained by multiplying the integral coefficient by the error at the current time. The derivative term is the value obtained by multiplying the derivative coefficient by the error at the current time, subtracting twice the error at the previous time, and adding the error at the previous two times.

6. The method for optimizing nonwoven fabric processing based on production data according to claim 1, characterized in that, The production rate control includes: The parameters to be controlled and the preset production parameters are obtained and processed to obtain a set of differences; all the parameters to be controlled are prioritized according to the differences in the set of differences to obtain the prioritized parameters to be controlled; the prioritized parameters to be controlled are obtained according to the priority and controlled according to the improved PID algorithm.

7. The method for optimizing nonwoven fabric processing based on production data according to claim 1, characterized in that, The rate regulation failure in step S4 includes: The first parameter control duration is obtained, which represents the complete time for the parameter to be controlled to be controlled according to the improved PID algorithm. If the first parameter control duration is longer than the preset first parameter control duration, it indicates that the rate adjustment has failed and a production safety warning is issued.

8. The method for optimizing nonwoven fabric processing based on production data according to claim 1, characterized in that, The phased rate control in step S5 includes: The parameters to be controlled and the preset production parameters are acquired and processed to obtain the first rate increment to be controlled; the first rate increment to be controlled is divided into N segments to obtain the second rate increment to be controlled; the second rate increment to be controlled is acquired and rate control is performed according to the improved PID algorithm. The time of the phased rate control is obtained and processed to obtain the second parameter control duration. If the second parameter control duration is longer than the preset second parameter control duration, it indicates that the rate adjustment in step S5 has failed, and a production safety warning is issued.

9. The method for optimizing nonwoven fabric processing based on production data according to claim 1, characterized in that, The production safety early warning in steps S4 and S5 is to obtain the control parameters and data that failed to be adjusted, and to warn of the processing steps and locations where production problems exist.

10. A nonwoven fabric processing optimization system based on production data, characterized in that, include: The nonwoven fabric quality analysis module is used to acquire nonwoven fabric and perform quality analysis to obtain the nonwoven fabric quality index. If the nonwoven fabric quality index is greater than the preset nonwoven fabric quality index threshold, it indicates that the production is normal. If it is not greater than the nonwoven fabric quality index threshold, it indicates that there is a problem in the production and the process will proceed to the nonwoven fabric production parameter acquisition module. The nonwoven fabric production parameter acquisition module is used to acquire production data during the nonwoven fabric production process. The production data includes: melt extrusion rate, spinning rate, airflow stretching rate, and collection net rate. The difference index calculation module is used to acquire and process the production data and preset production parameters to obtain the difference index, and divide the difference index into first-level difference, second-level difference and third-level difference according to the difference index classification function. The first-level difference processing module is used to acquire the production data and the preset production parameters and control the production rate through an improved PID algorithm; if the rate adjustment fails, a production safety warning is issued. The secondary difference processing module is used to acquire and process the production data to obtain the rate increment to be controlled, and to perform staged rate control through the improved PID algorithm; if the rate adjustment fails, a production safety warning is issued. The Level 3 Difference Processing Module is used to issue warnings that require production to be stopped and machines to be inspected.

Images